• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 4 answers
  • Subscribers 31 subscribers
  • Views 944 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TMP432: Asking for the TMP432 design questions

SHH
Mastermind 30251 points
Part Number: TMP432
Other Parts Discussed in Thread: TMP464, TMP435, TMP468

Hi Team,
Customer would like to use three TMP432 on the new design. Please help for this questions.
1. It looks like one TMPS432 number corresponds with one I2C slave address. If cusotmer need two TMP432 on the same bus, cusotmer have to use two part number(TMP432A/B), right?
2. In cusotmer design, customer need to ues 3 pcs*TMP432. But only two slave address can be selected. How can customer deal with it? Do TI have other part that has more address can be selectable?


Thanks,
SHH

  • 0
    TI__Mastermind 35855 points
    Hi SHH,

    1. Yes, customer would need to use 432A and 432B on the same bus.
    2. Why did customer select TMP432? Do they need Low Beta Correction? We have TMP468 which offers more channels than three TMP432 in one device. Customer can also use six TMP435 devices with Beta Correction feature. It seems unlikely that customer would need Beta Correction on all six channels, so I would recommend some combination of TMP432/5 and TMP464/8. If you have more information on the customer's remote junction, I can advise on Beta Correction.

    Thanks,
    Ren
  • 0 in reply to Ren Schackmann
    TI__Mastermind 30251 points
    Hi Ren,
    Could you advise how long the remote sensor(BJT) can be routed from TMP468?


    Thanks,
    SHH
  • 0 in reply to SHH
    TI__Mastermind 35855 points
    Hi SHH,

    Since we cancel series resistance and can tolerate a couple nF, the cable length isn't going to be limited by the properties of the wire. The biggest problem that we see in all remote junction applications is noise induced in the diode lines. The junction itself is relatively high impedance, and we pass only a few uA through it. In our lab, we have worked with cable lengths up to a meter, but we always use wiring with braided shielding. I recommend that all customers shield the diode traces if at all possible.

    Thanks,
    Ren
  • 0 in reply to Ren Schackmann
    TI__Mastermind 30251 points
    Hi Ren,
    thanks for your help!
    could you please help review the TMP468 schematic?
    txn.box.com/.../eq0fxebvaj48475scxyy503dy3qcgvto

    BR,
    SHH
  • 0 in reply to SHH
    TI__Mastermind 35855 points
    Hi SHH,

    This looks good, with the exception of the address connection. At a minimum, customer will need to not populate either R2603 or R2604. This pin is capable of configuring four different I2C addresses depending on whether it is tied to Supply, Ground, SCL or SDA. As drawn, it will be held at mid-supply, and give unpredictable address.

    Thanks,
    Ren
  • 0 in reply to Ren Schackmann
    TI__Mastermind 30251 points

    Hi Ren,

    Customer have one more question. In the spec, the RS and CDIFF are optional in most application. customer have no idea what value is better for our design. Could you help to advise? Or customer can keep current value?

    BR,

    SHH

  • 0 in reply to SHH
    TI__Mastermind 35855 points
    Hi SHH,

    Sorry for the delay; we had a TI holiday in the US on Friday.

    I don't have access to the customer schematic anymore, so I can't comment on current value.

    They are optional. We don't use them in our lab. The device has some internal filtering. Two 500ohm resistors, for a loop total of 1kOhm, and 100pF are typical values.

    Thanks,
    Ren